Chinese Chemical Letters  2019, Vol. 30 Issue (8): 1530-1532   PDF    
How long a C–C bond can be? An example of extraordinary long C-C single bond in 1, 2-diarylamino-o-carboranes
Yile Wu, Jie Zhang, Zuowei Xie*     
Department of Chemistry and State Key Laboratory of Synthetic Chemistry, The Chinese University of Hong Kong, Hong Kong, China
Abstract: How long a C-C bond can be? A question has long fascinated chemists. This work reports an example of extraordinary long C-C bond distance of 1.990(4) Å observed in single-crystal X-ray structure of 1, 2-(NHMes)2-o-carborane (2; Mes=2, 4, 6-trimethylphenyl). DFT calculations show that hyperconjugation of lone pairs of the nitrogen atoms into the empty σ* orbital of the cage C-C bond is the origin of the bond elongation. Such hyperconjugation can be suppressed if the two nitrogen atoms in 2 are linked to a Lewis acidic germanium (Ⅱ) center.
Keywords: o-Carborane     C–C single bond     Hyperconjugation     Amine     X-ray structure    

The C-C bonds are the basic building blocks in organic compounds and the carbon-based living things. Generally, the normal Csp3 - Csp3 bond length in organic compounds is about 1.54 Å [1]. However, C-C single bonds with deviant bond lengths have received much attention over the years [2-4]. The linear relationship between C-C bond length and the bond dissociation energies (BDE) reveals that the deviation of C-C bond lengths from the standard will cause significant changes in BDE [5]. Recently, substantial progress has been made in synthesizing compounds with elongated C-C bond lengths (Fig. 1). For example, hexaphenylethane (HPE) derivative A, reported by Mislow et al., features a long C-C bond length of 1.67(3) Å [6]. Other HPE derivatives B [7, 8] and C [9], prepared by the Toda and Herges groups, are proved to be stable compounds with long C-C bond lengths of 1.734(5) Å and 1.713(2) Å, respectively. Beyond HPE derivatives, Schreiner et al. demonstrated that the caged-alkane dimer D, with a C-C bond length elongated to 1.704(4) Å, is stabilized by attractive dispersion interactions [10]. Very recently, Suzuki et al. reported a series of dihydropyracylene with two spiro (dibenzocycloheptatriene) E having ultra-long C-C single bonds up to 1.806(2) Å in length [11].

Download:
Fig. 1. Reported compounds with elongated C-C bonds.

For 1, 2-disubstituted o-carboraness [12], previous reports suggest a possibility of elongating the cage C-C bonds by replacing the substituents on the cage carbon atoms [13, 14]. Both experimental [15, 16] and theoretical [17] results revealed that the aminosubstituting groups significantly elongated the corresponding cage C-C bonds due to the π-electron back-donation from the nitrogen lone pairs to the CC cage antibonding orbital. Especially, 1, 2-(NH2)2- 1, 2-C2B10H10 was predicted by DFT calculations to have longer C-C bonds than its –SH, OH, CH3 and PH2 derivatives [17]. However, the known method for synthesizing bisamino-o-carboraness was not known till very recently [18], and only mono-amino-substituted o-carboranes have been reported [19]. As an on-going project, we strive to continue with our effects in developing new methodologies for efficient functionalization of o-carboranes [19-22]. We report herein a novel one-pot synthesis of 1, 2-bis(arylamino)-o-carboranes that features the ultra long C-C single bond distance of 1.990(4) Å. While this paper was under preparation, the synthesis and structure of 1, 2-(NH2)2-o-carboranes and its derivatives were reported [18].

Inspired by the synthesis of arylamines from the reaction of aryl Grignard reagents with nitrosoarenes [23], we hypothesized the preparation of 1, 2-bis(arylamino)-o-carboraness directly from o-carboranes via a one-pot addition-reduction process. Hence, we started our investigation by reaction of 1, 2-(MgCl)2-o-C2B10H10 [24] with PhNO in THF, followed by treatment with LiBH4 and FeCl2, respectively. The desired product 1, 2-(phenylamino)2-o-C2B10H10 1 was isolated, after workup, in 26% yield. Replacement of 1, 2-(MgCl)2-o-C2B10H10 with 1, 2-dilitho-o-carboranes resulted in the isolation of 1 in 68% yield (Scheme 1).

Download:
Scheme 1. One-pot preparation of 1, 2-bis(arylamino)-o-carboranes.

Under the same manner, 1, 2-(NHMes)2-o-carboranes (2, Mes = 2, 4, 6-trimethylphenyl) was prepared in 58% yield (see the Supporting information for detail) (Scheme 1).

Both 1 and 2 were characterized by 1H, 13C and 11B NMR spectroscopy as well as HRMS. Their molecular structures are confirmed by single-crystal X-ray analyses (Fig. 2). Their key structure parameters together with other know analogues are compiled in Table 1. The most interesting structural feature is the especially long cage C-C bonds, 1.922(5) Å in 1 and 1.990(4) Å in 2. The latter is even longer than that of 1.931(2) Å observed in 1, 2- (MesCH2NH)2-o-C2B10H10 (F in Fig. 1) [18]. The average cage C-N bond distances of 1.380(4) Å in 1 and 1.388(3) Å in 2 are very close to that of 1.383(2) Å in F. The average B-B distances of 1.777 Å in 2, 1.774 Å in 3 and 1.777 Å in F are very comparable to each other.

Download:
Fig. 2. Molecular structures of 1 and 2 with the anisotropic displacement parameters depicted at the 30% and the 50% probability level, respectively. The H atoms are omitted for clarity.

Table 1
Key structural parameters for o-carborane, 2, 3 and F.

To gain more insight into the electronic structure of 2, DFT calculations were carried out for the structural optimization of o-carborane and 2 at the B3LYP/6-31G** level of theory [17]. Natural bond orbital (NBO) analyses (B3LYP/6-311G++**) indicate that the Wiberg bond index (WBI) of the cage C-C bond in 2 (0.33) is significantly smaller than that in o-carboranes (0.74), and is slightly smaller than that in F (0.34, Table 1). On the other hand, the WBI of 1.10 for Ccage–N bond in 2 suggests its multiple bonding characters. For the two nitrogen atoms in 2, the occupancies of long pairs (LPs) drop to a low value of 1.81e. Meanwhile, the empty σ* C-C orbital occupancy of 2 is calculated to be 0.44e, which is significantly larger than that of 0.07e in o-carboranes, indicating an electron transfer from the N atom to the empty σ* cage C-C orbital. These results are in line with the literature reports [17, 18], implying that the origin of the elongated cage C-C bonds in 1 and 2 results from the hyperconjugation between the LPs of the nitrogen atoms and the empty σ* C-C orbital, as illustrated in Scheme 2.

Download:
Scheme 2. Illustration of the hyperconjugation.

Furthermore, the Ccage–Ccage–N–Caryl torsion angles of 116.4(2)° in 2 are closer to an ideal torsion angel of 120° required to establish the parallelism between the LP orbital of nitrogen and σ* C-C orbital. In contrast, such torsion angles in 1 [83.9(4) to 87.9(4)°] and F [108.1(2) and 96.0(2)°] (Table 1) are largely off the ideal one. These results reveal that the observed parallelism between LP orbital of nitrogen and σ* C-C orbital in 2 enhances the hyperconjugation, leading to an extraordinary long C-C bond.

It is clear that the lengthening of the cage C-C bond is closely related to the hyperconjugation aforementioned. We anticipated that such hyperconjugation would be suppressed if both N atoms in 2 are bonded to a strong Lewis acidic center. In this connection, treatment of 2 with 2 equivalent of n-BuLi in toluene, followed by the addition of 1 equiv. of GeCl2·dioxane, gave a germylene 3 in 81% yield (Scheme 3). Compound 3 was fully characterized by 1H, 13C, 11B NMR and HRMS as well as single-crystal X-ray analysis.

Download:
Scheme 3. Synthesis and structure of germylene 3. The anisotropic displacement parameter is depicted at the 50% probability level. The H atoms are omitted for clarity.

The cage C-C bond distance of 1.637(3) Å in 3 is close to that of 1.624(8) Å in o-carboranes, but is significantly shorter than that of 1.990(4) Å observed in 2 (Table 1). Accordingly, the Ccage–N bond distance of 1.414(3) Å in 3 is much longer than that of 1.388(3) Å in 2. NBO analyses (at the B3LYP/6-311G++** level of theory) predict a WBI of 0.75 for the cage C-C bond in 3, which is very similar to that of 0.74 in o-carboranes (Table 1). On the other hand, a WBI of 1.03 for the Ccage–N bond in 3 is considerably smaller than that of 1.10 in 2. These results suggest that no hyperconjugation is observed between LPs of the nitrogen atoms and the σ* orbital of the cage C-C bond in 3. Instead, the nitrogen atoms donate their LPs to the empty p-orbital of the Ge(Ⅱ) center in 3, stabilizing such a germylene compound [26].

In summary, we describe a new one-pot synthesis of 1, 2-bis (arylamino)-o-carboraness 1 and 2, by treatment of 1, 2-dilithio-o-carborane with nitrosoarenes. Single-crystal X-ray analyses reveal that 1 and 2 have extraordinary long cage C-C bonds with distances of 1.922(5) Å and 1.990(4) Å, respectively. On the other hand, a WBI of 1.10 for Ccage–N in 2 suggests its multiple bonding characters. DFT calculations on 2 indicate that the hyperconjugation between the LPs of nitrogen atoms and the empty σ* C-C orbital is responsible for the elongated C-C bonds in 1 and 2.

How long a C-C bond can be? Though this is still an open question, the current work pushes the limit further out.

Acknowledgments

This work was supported by a grant from the Research Grants Council of the Hong Kong Special Administration Region (No. 14305918).

Appendix A. Supplementary data

Supplementary material related tothis article can befound, in the online version, at doi:https://doi.org/10.1016/j.cclet.2019.04.034.

References
[1]
F.H. Allen, O. Kennard, D.G. Watson, et al., J. Chem. Soc. Perkin Trans. I 2 (1987) S1-S19.
[2]
S. Vázquez, P. Camps, Tetrahedron 61 (2005) 5147-5208. DOI:10.1016/j.tet.2005.03.055
[3]
J.M. McBride, Tetrahedron 30 (1974) 2009-2022. DOI:10.1016/S0040-4020(01)97332-6
[4]
M. Tanaka, A. Sekiguchi, Angew. Chem. Int. Ed. 44 (2005) 5821-5823. DOI:10.1002/anie.200501605
[5]
A.A. Zavitsas, J. Phys. Chem. A 107 (2003) 897-898. DOI:10.1021/jp0269367
[6]
B. Kahr, D. Van Engen, K. Mislow, J. Am. Chem. Soc. 108 (1986) 8305-8307. DOI:10.1021/ja00286a053
[7]
K. Tanaka, N. Takamoto, Y. Tezuka, M. Kato, F. Toda, Tetrahedron 57 (2001) 3761-3767. DOI:10.1016/S0040-4020(01)00249-6
[8]
F. Toda, K. Tanaka, M. Watanabe, et al., J. Org. Chem. 64 (1999) 3102-3105. DOI:10.1021/jo982010c
[9]
S. Kammermeier, R. Herges, P.G. Jones, Angew. Chem. Int. Ed. 36 (1997) 1757-1760. DOI:10.1002/anie.199717571
[10]
P.R. Schreiner, L.V. Chernish, P.A. Gunchenko, et al., Nature 477 (2011) 308. DOI:10.1038/nature10367
[11]
Y. Ishigaki, T. Shimajiri, T. Takeda, R. Katoono, T. Suzuki, Chemistry 4 (2018) 795-806. DOI:10.1016/j.chempr.2018.01.011
[12]
R.N. Grimes, Icosahedral carboranes: 1, 2-C2B10H12, in: R.N. Grimes (Ed.), Carboranes, third edition, Academic Press, Boston, 2016, pp. 283-502.
[13]
J. Llop, C. Viñas, J.M. Oliva, et al., J. Organomet. Chem. 657 (2002) 232-238. DOI:10.1016/S0022-328X(02)01637-6
[14]
R.A. Harder, J.A. HughMacBride, G.P. Rivers, et al., Tetrahedron 70 (2014) 5182-5189. DOI:10.1016/j.tet.2014.05.102
[15]
L.A. Boyd, W. Clegg, R.C.B. Copley, et al., Dalton Trans. (2004) 2786-2799.
[16]
M.A. Fox, R.J. Peace, W. Clegg, Polyhedron 28 (2009) 2359-2370. DOI:10.1016/j.poly.2009.04.041
[17]
J.M. Oliva, N.L. Allan, Pv.R. Schleyer, C. Viñas, F. Teixidor, J. Am. Chem. Soc. 127 (2005) 13538-13547. DOI:10.1021/ja052091b
[18]
J. Li, R. Pang, Z. Li, et al., Angew. Chem. Int. Ed. 58 (2019) 1397-1401. DOI:10.1002/anie.201812555
[19]
R. Cheng, J. Zhang, J. Zhang, Z. Qiu, Z. Xie, Angew.Chem.Int.Ed. 55 (2016) 1751-1754. DOI:10.1002/anie.201507952
[20]
Z. Qiu, S. Ren, Z. Xie, Acc. Chem. Res. 44 (2011) 299-309. DOI:10.1021/ar100156f
[21]
J. Zhang, Z. Xie, Acc. Chem. Res. 47 (2014) 1623-1633. DOI:10.1021/ar500091h
[22]
Y. Quan, Z. Qiu, Z. Xie, Chem. -Eur. J. 24 (2018) 2795-2805. DOI:10.1002/chem.201704937
[23]
I. Sapountzis, P. Knochel, J. Am. Chem. Soc. 124 (2002) 9390-9391. DOI:10.1021/ja026718r
[24]
C. Tang, Z. Xie, Angew. Chem. Int. Ed. 54 (2015) 7662-7665. DOI:10.1002/anie.201502502
[25]
M.J. Hardie, C.L. Raston, Crystengcomm 3 (2001) 162-164. DOI:10.1039/b107198j
[26]
Y. Mizuhata, T. Sasamori, N. Tokitoh, Chem. Rev. 109 (2009) 3479-3511. DOI:10.1021/cr900093s